Hadrian OpenHack Finds Bugs Better with AI Pentesting

Hadrian released OpenHack, an open-source, human-in-the-loop platform that automates white‑box source code review and penetration testing by orchestrating focused LLM-driven sub-agents for reconnaissance, scenario routing, exploit construction, and triage; the article notes Hadrian previously found hundreds of vulnerabilities in government-used open source software — including a critical flaw that exposed Azure database credentials — and contrasts OpenHack's scoped approach with large LLMs that lack persistent context and scope management.